From 35ca26c898bcffe6e3b5c930d64a8988bd91f2a0 Mon Sep 17 00:00:00 2001 From: Daniel Han Date: Wed, 4 Dec 2024 05:38:05 -0800 Subject: [PATCH] Dynamic quants (#1379) * typing * Update cross_entropy_loss.py * Update cross_entropy_loss.py * Update cross_entropy_loss.py * Update cross_entropy_loss.py * Update cross_entropy_loss.py * Update cross_entropy_loss.py * Update cross_entropy_loss.py * Update cross_entropy_loss.py * Update cross_entropy_loss.py * int64 * Update _utils.py * Update cross_entropy_loss.py * constexpr * constexpr * Update cross_entropy_loss.py * Update cross_entropy_loss.py * Update _utils.py * Update _utils.py * Update _utils.py * CE * Update cross_entropy_loss.py * Update _utils.py * Update llama.py * Update _utils.py * Update rms_layernorm.py * Update rms_layernorm.py * Update rms_layernorm.py * Update rms_layernorm.py * Update rms_layernorm.py * Update rms_layernorm.py * Update utils.py * Update rms_layernorm.py * Update rms_layernorm.py * Update rms_layernorm.py * Update rms_layernorm.py * Update rms_layernorm.py * Update rms_layernorm.py * Update rms_layernorm.py * Update rms_layernorm.py * Update rms_layernorm.py * Update rms_layernorm.py * Update rms_layernorm.py * Update rms_layernorm.py * typing * Update rope_embedding.py * types * Disable compiling * Update _utils.py * Update _utils.py * Forward hook * Update _utils.py * Update llama.py * Update _utils.py * Update llama.py * Update llama.py * Update _utils.py * Update pyproject.toml * Update _utils.py * Update llama.py * CE Loss * Update cross_entropy_loss.py * Update _utils.py * Update cross_entropy_loss.py * Update cross_entropy_loss.py * Update cross_entropy_loss.py * Update llama.py * Update _utils.py * Update _utils.py * Update _utils.py * Update _utils.py * Update _utils.py * Fix: cast logits to float32 in cross_entropy_forward to prevent errors (#1254) * Fix: cast logits to float32 in cross_entropy_forward to prevent errors * Update cross_entropy_loss.py --------- Co-authored-by: Daniel Han * Throw error when inferencing longer than max_popsition_embeddings (#1236) * Throw error when inferencing longer than max_popsition_embeddings without rope scaling * Update llama.py --------- Co-authored-by: Daniel Han * CLI now handles user input strings for dtype correctly (#1235) Co-authored-by: root * Update flex_attention.py * Update _utils.py * Update _utils.py * Update flex_attention.py * Update flex_attention.py * Update loader.py * Update loader.py * Update flex_attention.py * Update flex_attention.py * Update flex_attention.py * Update flex_attention.py * Update _utils.py * Update cross_entropy_loss.py * Update _utils.py * Update tokenizer_utils.py * Update tokenizer_utils.py * Update tokenizer_utils.py * Update tokenizer_utils.py * Update tokenizer_utils.py * triton_cast * Update utils.py * Qwen 2.5 Coder * Fix/export mistral (#1281) * Enhance install_python_non_blocking to handle protobuf installation and process management * Revert "Enhance install_python_non_blocking to handle protobuf installation and process management" This reverts commit a3b796a05841fb8d93c652c845591e12cf81ea93. * Set PROTOCOL_BUFFERS_PYTHON_IMPLEMENTATION to 'python' to address issue #1266 * Revert "Set PROTOCOL_BUFFERS_PYTHON_IMPLEMENTATION to 'python' to address issue #1266" This reverts commit f00fbf5eac7ad4f5d48c70b98d770255d1a9ef58. * Set PROTOCOL_BUFFERS_PYTHON_IMPLEMENTATION to 'python' to address issue #1266 * Update __init__.py --------- Co-authored-by: Daniel Han * DOC Update - Update README.md with os.environ in example (#1269) * Update README.md with os.environ in example Added OS Environ in example to avoid device conflicts , for a user at least in jupyter notebook this allows to select GPU in a multi GPU setup. As currently the unsloth init checks all GPU's and takes the first in the order which can be a issue when some GPU's are in use and the list still shows them. So to manually avoid this, this os config is required. Small change but a bit time saver for those who straight away copies the tutorials * Update README.md --------- Co-authored-by: Daniel Han * fix/get_chat_template (#1246) * Refactor `get_chat_template` to now support system message instead. It supposed to fix ollama tokenizer chattemplate to * Remove type hinting * Update chat_templates.py --------- Co-authored-by: Daniel Han * fix/sft-trainer (#1276) * Add patch for SFTTrainer to maintain backward compatibility with TRL changes * Update trainer.py * Update trainer.py * Refactor trainer patch to maintain backward compatibility with TRL changes * Update trainer.py * Refactor trainer.py to exclude non-convertible trainers from backward compatibility patch --------- Co-authored-by: Daniel Han * Update __init__.py * Update trainer.py * Update trainer.py * Update trainer.py * Update tokenizer_utils.py * Update llama.py * Fix #853 * fix/sfttrainer-compatibility (#1293) * Refactor trainer.py to import SFTConfig directly and update UnslothTrainingArguments class inheritance * Update trainer.py * Update trainer.py --------- Co-authored-by: Daniel Han * Update rms_layernorm.py * Update rms_layernorm.py * Gemma * Update rms_layernorm.py * Update gemma2.py * Cut Cross Entropy * Update llama.py * Cut Cross Entropy * Update llama.py * Update llama.py * Update llama.py * Update __init__.py * Update __init__.py * Update _utils.py * Update _utils.py * Update _utils.py * Update _utils.py * Update _utils.py * Update _utils.py * Update _utils.py * Update _utils.py * Update mapper.py * Update _utils.py * Update _utils.py * Update _utils.py * Update _utils.py * Update _utils.py * Update _utils.py * Update _utils.py * Update _utils.py * Update _utils.py * patch_fast_lora * vision * Update fast_lora.py * Update _utils.py * Update _utils.py * Vision * Update trainer.py * Update save.py * FastBaseVisionModel * Update loader_utils.py * Update vision.py * Update loader.py * Update vision.py * Update loader.py * Update vision.py * Update _utils.py * tokenizer_name * Update loader.py * Update vision.py * Update save.py * Update save.py * Update vision.py * Update vision.py * Update vision.py * Update vision.py * Update vision.py * Update vision.py * Update _utils.py * Update loader.py * kwargs * logits * Update llama.py * Update llama.py * Update llama.py * Update _utils.py * Update _utils.py * Update _utils.py * error * Update _utils.py * Update _utils.py * Update _utils.py * Update _utils.py * Update _utils.py * Update _utils.py * Update _utils.py * Update _utils.py * Update _utils.py * Update _utils.py * Update _utils.py * Update _utils.py * Update _utils.py * Update _utils.py * Update _utils.py * Update loader.py * Update llama.py * Update vision.py * Update loader.py * Old torch versions * Update loader.py * Update loader.py * prints * recheck * Update loader.py * Update loader.py * Update _utils.py * Update _utils.py * Update mapper.py * Feat/kto (#1316) * Add PatchKTOTrainer and update model imports * Update dpo.py * Update __init__.py * Delete unsloth/models/kto.py --------- Co-authored-by: Daniel Han * Fix orpo/dpo trainer (#1286) * change the colab notebook for dpo zephyr and orpo * use original tokenizer * Update README.md * Update README.md --------- Co-authored-by: Daniel Han * skip modules * Update vision.py * Update llama.py * Update llama.py * Update llama.py * Update llama.py * Update llama.py * Update llama.py * Update llama.py * Update llama.py * Update llama.py * Update llama.py * Update llama.py * Fix llama.cpp * Update save.py * Update save.py * Update vision.py * Update save.py * Update save.py * Update save.py * Update save.py * Update save.py * Update save.py * Update save.py * Update _utils.py * Update save.py * Update save.py * Update mapper.py * modules --------- Co-authored-by: Edd <68678137+Erland366@users.noreply.github.com> Co-authored-by: Datta Nimmaturi Co-authored-by: Edwin Fennell Co-authored-by: root Co-authored-by: Uday Girish Maradana Co-authored-by: cell-dame <122996026+dame-cell@users.noreply.github.com> --- unsloth/models/_utils.py | 2 +- unsloth/models/mapper.py | 38 +++++++++++++++++++++++++++++++------- unsloth/models/vision.py | 4 ++-- 3 files changed, 34 insertions(+), 10 deletions(-) diff --git a/unsloth/models/_utils.py b/unsloth/models/_utils.py index 5bc9529e60..11423f9066 100644 --- a/unsloth/models/_utils.py +++ b/unsloth/models/_utils.py @@ -12,7 +12,7 @@ # See the License for the specific language governing permissions and # limitations under the License. -__version__ = "2024.12.1" +__version__ = "2024.12.2" __all__ = [ "prepare_model_for_kbit_training", diff --git a/unsloth/models/mapper.py b/unsloth/models/mapper.py index b2f73aa6c2..0ba03ce01c 100644 --- a/unsloth/models/mapper.py +++ b/unsloth/models/mapper.py @@ -452,33 +452,45 @@ __INT_TO_FLOAT_MAPPER = \ "unsloth/Llama-3.1-Nemotron-70B-Instruct", "nvidia/Llama-3.1-Nemotron-70B-Instruct-HF", ), - "unsloth/Qwen2-VL-2B-Instruct-bnb-4bit" : ( + "unsloth/Qwen2-VL-2B-Instruct-unsloth-bnb-4bit" : ( "unsloth/Qwen2-VL-2B-Instruct", "Qwen/Qwen2-VL-2B-Instruct", + "unsloth/Qwen2-VL-2B-Instruct-bnb-4bit", ), - "unsloth/Qwen2-VL-7B-Instruct-bnb-4bit" : ( + "unsloth/Qwen2-VL-7B-Instruct-unsloth-bnb-4bit" : ( "unsloth/Qwen2-VL-7B-Instruct", "Qwen/Qwen2-VL-7B-Instruct", + "unsloth/Qwen2-VL-7B-Instruct-bnb-4bit", ), - "unsloth/Llama-3.2-11B-Vision-Instruct-bnb-4bit" : ( + "unsloth/Qwen2-VL-72B-Instruct-unsloth-bnb-4bit" : ( + "unsloth/Qwen2-VL-72B-Instruct", + "Qwen/Qwen2-VL-72B-Instruct", + "unsloth/Qwen2-VL-72B-Instruct-bnb-4bit", + ), + "unsloth/Llama-3.2-11B-Vision-Instruct-unsloth-bnb-4bit" : ( "unsloth/Llama-3.2-11B-Vision-Instruct", "meta-llama/Llama-3.2-11B-Vision-Instruct", + "unsloth/Llama-3.2-11B-Vision-Instruct-bnb-4bit", ), - "unsloth/Llama-3.2-90B-Vision-Instruct-bnb-4bit" : ( + "unsloth/Llama-3.2-90B-Vision-Instruct-unsloth-bnb-4bit" : ( "unsloth/Llama-3.2-90B-Vision-Instruct", "meta-llama/Llama-3.2-90B-Vision-Instruct", + "unsloth/Llama-3.2-90B-Vision-Instruct-bnb-4bit", ), - "unsloth/Llama-3.2-11B-Vision-bnb-4bit" : ( + "unsloth/Llama-3.2-11B-Vision-unsloth-bnb-4bit" : ( "unsloth/Llama-3.2-11B-Vision", "meta-llama/Llama-3.2-11B-Vision", + "unsloth/Llama-3.2-11B-Vision-bnb-4bit", ), - "unsloth/Llama-3.2-90B-Vision-bnb-4bit" : ( + "unsloth/Llama-3.2-90B-Vision-unsloth-bnb-4bit" : ( "unsloth/Llama-3.2-90B-Vision", "meta-llama/Llama-3.2-90B-Vision", + "unsloth/Llama-3.2-90B-Vision-bnb-4bit", ), - "unsloth/Pixtral-12B-2409-bnb-4bit" : ( + "unsloth/Pixtral-12B-2409-unsloth-bnb-4bit" : ( "unsloth/Pixtral-12B-2409", "mistralai/Pixtral-12B-2409", + "unsloth/Pixtral-12B-2409-bnb-4bit", ), "unsloth/Pixtral-12B-2409-Base-bnb-4bit" : ( "unsloth/Pixtral-12B-Base-2409", @@ -500,6 +512,10 @@ __INT_TO_FLOAT_MAPPER = \ "unsloth/Llama-3.1-Tulu-3-70B", "allenai/Llama-3.1-Tulu-3-70B", ), + "unsloth/QwQ-32B-Preview-bnb-4bit" : ( + "unsloth/QwQ-32B-Preview", + "Qwen/QwQ-32B-Preview", + ), } INT_TO_FLOAT_MAPPER = {} @@ -519,6 +535,14 @@ for key, values in __INT_TO_FLOAT_MAPPER.items(): MAP_TO_UNSLOTH_16bit[values[1]] = values[0] MAP_TO_UNSLOTH_16bit[values[1].lower()] = values[0] pass + elif len(values) == 3: + # Dynamic Unsloth quantization + if values[0].startswith("unsloth"): + MAP_TO_UNSLOTH_16bit[values[1]] = values[0] + MAP_TO_UNSLOTH_16bit[values[1].lower()] = values[0] + MAP_TO_UNSLOTH_16bit[values[2]] = values[0] + MAP_TO_UNSLOTH_16bit[values[2].lower()] = values[0] + pass pass # Get lowercased diff --git a/unsloth/models/vision.py b/unsloth/models/vision.py index aa4cc09022..0f682c6b40 100644 --- a/unsloth/models/vision.py +++ b/unsloth/models/vision.py @@ -30,7 +30,7 @@ from transformers import set_seed as transformers_set_seed from unsloth_zoo.peft_utils import ( get_peft_regex, merge_and_overwrite_lora, - # SKIP_QUANTIZATION_MODULES, + SKIP_QUANTIZATION_MODULES, ) from triton import __version__ as triton_version @@ -133,7 +133,7 @@ class FastBaseVisionModel: bnb_4bit_use_double_quant = True, bnb_4bit_quant_type = "nf4", bnb_4bit_compute_dtype = dtype, - # llm_int8_skip_modules = SKIP_QUANTIZATION_MODULES, + llm_int8_skip_modules = SKIP_QUANTIZATION_MODULES, ) pass